Biostatistician & R Developer

I build open-source R software and lightweight web tools for reproducible analysis of biomedical data—from cheminformatics and structural bioinformatics to clinical analytics.

My publications span R software for statistical modeling and programmatic access to chemical/structural data; interactive tools for survival analysis, biological variation, and virtual screening; and methods studies on deep learning for small-molecule classification and imbalanced datasets in drug discovery. Recent essays examine statistical rigor and reproducibility in the AI era and the promise and limits of AI in healthcare. I’m currently developing fastml, an R package for fast model training and evaluation with streamlined preprocessing, multi-algorithm comparison, hyperparameter tuning, and clear performance visualizations; and StatGuide, an AI-assisted manuscript checker aligned with CONSORT, ARRIVE, STROBE, and related reporting standards. My focus is principled evaluation under class imbalance and limited samples, transparent reporting, and end-to-end, auditable R workflows in close collaboration with clinicians and life scientists.